Gut Supplement, 1991, S73-S78

S73

Primary biliary cirrhosis J

Neuberger, M Lombard, R Galbraith

During the last 25 years there have been considerable advances in the understanding of primary biliary cirrhosis. It is now appreciated that the disease has a much wider clinical spectrum than previously thought, with many people being diagnosed in an asymptomatic state. The increasing success of liver transplantation has resulted in a greater need for accurate assessment of prognosis. Development of accurate prognostic models to assess survival has been aided by multicentre studies where large cohorts of patients have been followed for many years. Many immunological abnormalities associated with the disease have been identified: the diverse nature of the antimitochondrial antibodies - the serological hallmark ofthe disease - is recognised and the subtypes are being increasingly identified. The antigen, pyruvate dehydrogenase complex, has been defined and the antigenic epitopes are being defined. Yet, despite these major advances over the last quarter century, the pathogenesis ofprimary biliary cirrhosis remains elusive. Failure to define the pathogenesis of disease has resulted in a lack of a logical approach to therapy. Despite the increasing number of trials being carried out, no drug has yet shown to be totally effective in preventing progression of the disease.

The Liver and Hepatobiliary Unit, The Queen Elizabeth Hospital, Edgbaston, Bimingham J Neuberger

Department of Medicine, University of Liverpool, PO Box 147, Liverpool M Lombard Medical University of South Carolina, 171

Ashley Avenue, Charleston, South Carolina, USA R Galbraith Correspondence to: Dr J Neuberger.

Clinical features Perhaps one of the greater strengths of the Liver Unit in the field of primary biliary cirrhosis has been the ability to collect and monitor a large series of patients with a comparatively rare disease. There has been a longstanding collaboration between the Liver Unit at King's College Hospital, the Division of Hepatology in the Rigshospitalet, Copenhagen and the Liver Unit in the Hospital Clinic i Provincial, Barcelona. The first landmark paper describing the clinical course of primary biliary cirrhosis appeared in 1980' and was followed by a review of presenting symptoms five years later.2 Of the presenting symptoms, 47% presented with pruritus, 22% presented with non-specific symptoms, mainly tiredness, fatigue and right upper quadrant pain. Nineteen per cent presented with hepatic decompensation jaundice, gastrointestinal bleeding or ascites. In 5% the presenting symptoms were suggestive of vascular disease or auto immune disease (such as arthritis, scleroderma, or Raynaud's phenomenon). It was of interest that even at that time 7% were identified who presented with no specific hepatological symptoms. Modes of presentation of this asymptomatic group included abnormalities on routine blood testing, or during investigation for other diseases such as diabetes mellitus. Follow up studies suggested that during first five years after presentation, jaundice increased from 59% to 82% and hepatomegaly from 54% to 86%. -

Gastrointestinal bleeding increased from 17% to 46% and ascites from 7% to 49%. Cirrhosis increased more rapidly from 30% to 82% in four years. Analysis of risk factors at that time suggested that age, histological stage, and most importantly serum bilirubin were of major prognostic value. Although serum immunoglobulins and titres of antimitochondrial antibody tended to rise with time, and ascites and gastrointestinal bleeding became more common, these were of less prognostic importance. During the study 36 patients died and perhaps not surprisingly, over 75% of these patients had hepatomegaly, jaundice and on histology, peripheral cholestasis and cirrhosis. One other feature of note was the poor correlation of symptoms and biochemical abnormalities with histological staging. Although, as might be anticipated those with more advanced histological stage did have more evidence of decompensation and a higher serum bilirubin, of those with histological stage one (14% of the whole series) 39% had jaundice and 11% gastrointestinal bleeding; equally, 9% required treatment with diuretics. Despite the marked female preponderance of the disease (about 90%), the clinical course of disease in the two sexes appears similar.3 Although there was no difference in survival, it was observed that men present less commonly with pruritus and more commonly are detected on routine blood tests. While the latter may reflect merely the increased use of health screening in men, the former is of interest in particular when it was found that premenstrual women had more pruritus than postmenstrual women. The cause of pruritus in primary biliary cirrhosis is not understood; although recent work suggests that the endogenous opiate agonists ligands (such as encephalins) may be involved,4 the sex differences raise the possibility that sex steroids may modify response to pruritus. With the increasing use of auto analysers, the greater availability of auto antibody testing, and the generally greater awareness by pathologists and clinicians of the possibility of primary biliary cirrhosis, appreciation of the clinical spectrum has widened dramatically. The term 'asymptomatic primary biliary cirrhosis' was introduced by Fox in 1973 and achieved much greater publicity after the review by James et al.5 A large study formed the basis of a retrospective analysis of 95 patients with asymptomatic primary biliary cirrhosis in collaboration between the Liver Unit and Newcastle.6 The majority of these patients had early disease histologically assessed, but contrary to some reports, it was found these patients do progress clinically and histologically. Thirty five per cent of patients become symptomatic and the survival of these patients was no different from those who present with symptomatic disease and 15 of these died of liver related

S74

Neuberger, Lombard, Galbraith

causes. Mortality was significantly greater than the normal age and sex matched population. Thus, although asymptomatic disease may represent a more benign end of the spectrum, there can be little doubt that there is progression in many patients. Overall, there does appear to be a wide variation in the rate of progression between patients. Regarding side effects of the disease, bone disease can induce major morbidity and indeed mortality.7'8 Osteoporosis may result from either cortical or trabecular bone loss. The latter is more important as vertebral, radial, and femoral neck fractures may result. Stellon in conjunction with Juliette Compston studied this problem in detail. Of 36 patients with cholestasis, 42% had symptoms suggestive of osteoporosis and 11% had one or more wedged crushed vertebrae. Compared with age and sex matched controls, 28% had excessive cortical and 36% excessive trabecular bone loss. Not surprisingly, bone loss was greatest in those who had received corticosteroids. In contrast, in this and most other series, osteomalacia rarely occurs, and vitamin D deficiency is associated mainly with prolonged cholestasis. Further insight was gained by histological studies after double tetracycline labelling. The combination of increased osteoblastic function, decreased mineral apposition rate and prolonged mineralisation lag time suggest a defect in matrix synthesis. Additional studies suggested increase in resorption may also contribute to bone loss in primary biliary cirrhosis. The cause of this major complication remains uncertain. It may relate to the effects of cytokines which in vitro affect osteoblastic activity. Treatment remains uncertain: avoidance of corticosteroids, use of vitamin D supplements in association with calcium supplements and sodium fluoride have been advocated. There is no universal agreement for optimal treatment. Of the other features associated with primary biliary cirrhosis, thyroid disorders and its association with thyroid autoantibodies and the Sicca syndrome,9 scleroderma,'° the incidence and pattern of rheumatic disorders1' and the association with increasing levels of circulating immune complexes'2 and menstrual abnormalities'3 have been reported. Hepatocellular carcinoma was initially thought to be rare in this disease but a retrospective study of patients who developed primary biliary cirrhosis and hepatocellular carcinoma revealed that these patients are at risk of this complication.'4 The development of hepatocellular carcinoma was often insidious and associated with the general progression of the disease. Serum alphafetoprotein was often raised and was sometimes the only antimortem evidence of malignant transformation. The proTABLE I Problems with trials of therapy in primary biliary

cirrhosis Unknown aetiology Long natural history Uncertain progression Withdrawal for transplantation Choice of end point

portion of men to women developing hepatocellular carcinoma was similar to that in other forms of cirrhosis, emphasising work from the Unit and others, that duration of cirrhosis and male sex are the major risk factors for hepatoma development in United Kingdom.

Histological features The morphological and histological features of primary biliary cirrhosis have been well described. The initial, often focal lesion of chronic non-suppurative destructive cholangitis primarily affects bile ducts of 50-100 [tm diameter. The cellular infiltrate is often mixed but Tlymphocytes predominate. In the normal liver, biliary epithelium may express HLA class I antigens but rarely class II antigens. Increased class II expression may suggest that biliary epithelial can present antigen but there is little evidence as yet that this can occur. Increased class II expression occurs in other diseases, such as sclerosing cholangitis and graft versus host disease, and thus may be a secondary phenomenon. The availability of liver histology from the follow up patients allowed analysis of the sequential and diagnostic features of primary biliary cirrhosis. In association with Hans Popper, 605 liver biopsy specimens from 209 patients with primary biliary cirrhosis were analysed."5 Four varieties of piecemeal necrosis (biliary, lymphocytic, ductular, and fibrotic) were identified. Lymphocyte piecemeal necrosis, usually identified with lobular hepatitis, was found in histological stage 2 and 3, whereas biliary piecemeal necrosis was present in stages 2, 3, and 4. The fibrotic biliary piecemeal necrosis appeared only in the late cirrhotic stages. Although cholestatic features dominated, in 23% there were histological features compatible with overlap with chronic active hepatitis. These overlap features tended to be maintained throughout the course of the disease. When the biopsies of patients who died were analysed, features suggesting rapidly progressive disease included the development of cirrhosis, increasing cholestasis, especially in association with prominent hyaline inclusions, abundance of coarse collagen fibres and 'halo formation' at the margin of fibrous septa. The controversial report of an association of granulomas with worse prognostic disease was not supported by this study. Again the problems of the conventional staging and lack of correlation with clinical features were described. Therapy The development of logical strategies for therapy is dependent on knowledge of the pathogenesis of the disease (Table I). The relatively slow progression of the disease makes drug trials difficult. Ultimately, the endpoint of any study must be either death or transplantation. Even alteration of known prognostic variables such as serum bilirubin does not necessarily mean that the prognosis has been altered. The natural history of the disease, the variability in presentation and the development of complications such as ascites or variceal haemorrhage means

Primary biliary cirrhosis

S75

that large numbers of patients have to be studied and followed for many years. Indeed, based on an analysis of placebo patients in the various drug studies, it has been calculated that if a drug is associated with a 50% reduction in mortality, over 200 patients have to be followed for five years to show such an effect with statistical confidence. This has considerable logistic implications which will be fully appreciated by anyone who has been involved in drug studies! The many immunological abnormalities associated with the disease have suggested that immune mechanisms are important in the pathogenesis and perpetuation of the disease and indeed primary biliary cirrhosis has been suggested to be an autoimmune disease. Early studies with corticosteroids showed little effect and their use has not been advocated because of the risk of exacerbating osteoporosis. More recent studies, however, have suggested that serum biochemistry may be improved by use of corticosteroids but the effect on bones remains uncertain.'6 17 A multicentre prospective, placebo controlled double blind study was set up to study the effect of azathioprine in 248 patients recruited from Europe and followed for up to 10 years.'819 This study has proved a landmark for a number of reasons. Although the overall mortality in the two groups of patients was not significantly different, there was a slight but important imbalance between the two groups, in that those randomised to receive azathioprine had significantly more advanced disease. Correction for these imbalances suggested that azathioprine was associated with a small but significant improvement in survival (azathioprine reducing the mortality risk to 59% of that observed in the placebo group - 95% confidence intervals 40-90%). This would translate to an average improved survival of 20 months. Furthermore, azathioprine was associated with a reduction in the rate of symptomatic deterioration (Fig 1). An additional advantage from this study was the development of prognostic models for survival (see below). At the time the study was concluding, preliminary reports were appearing in the literature suggesting that D-penicillamine was associated with reduced mortality, especially in those patients treated with stage 3 and 4 disease. This 1 *0 08

O 0-6 Q.>

04 02*

w

. I

0 12 24 36 48 607284

-4

96108120

Time (months) AZA: 112 92 76 64 55 44 30 19 17 13 9 Noof PLAC: 104 84 71 59 47 30 18 11 5 4 0 patients Figure 1: Survival probability in patients with primary biliary cirrhosis in controls and receiving azathioprine (from ref 19).

TABLE II Specific therapyfor survival in primary biliary

cirrhosis Ineffective: Uncertain benefit: Limited benefit:

D-Penicillamine Corticosteroids Ursodeoxycholic acid Azathioprine Cyclosporin A

Colchicine Proven benefit:

Methotrexate Transplantation

prompted the start of a new study22 where D-penicillamine at a dose of 1200 mg/day was assessed. One hundred and eighty nine patients entered the study and were followed for up to four years. Analysis showed no significant difference in survival between the two groups and there were no significant differences in disease state between the two groups. Indeed the only significant difference observed in the study was the withdrawal due to drug related side effects (36% against 89%). The lack of response to D-penicillamine has also been supported by other studies.2324 In the last few years there have been reports of trials with a number of different agents (Table II). These include colchicine, methotrexate, chlorambucil, malotilate, ursodeoxycholic acid, and cyclosporin.25 3 Of these, chlorambucil was suggested to be effective but too toxic. Methotrexate has been suggested to return liver tests to normal in some cases, but there are, as yet, no longterm controlled studies. Ursodeoxycholic acid (UDCA) has achieved much notoriety. There is an overall consensus that the drug often relieves pruritus and may improve liver function tests; one study30 suggested that the biochemical improvement was only transient. Results are awaited of longterm placebo controlled studies. In none of these studies has there been any clear cut effect on survival although much emphasis has been placed on differences in serum bilirubin and immunoglobulins between the two groups. Whether a reduction in serum bilirubin is associated with improved mortality has yet to be shown. Cyclosporin corrects many of the immunological abnormalities in primary biliary cirrhosis,29 but a preliminary study using a dose of 10 mg/kg/day was discontinued because of nephrotoxicity. Smaller doses may be more effective.3233 The Liver Unit, continuing its well established collaboration with Barcelona and Copenhagen have just analysed a trial assessing cyclosporin 3 mg/kg/day.6 Three hundred and forty nine patients were followed for up to six years. Results were similar to the azathioprine study although survival was longer. Cox regression analysis showed cyclosporin prolongs survival to death and transplantation by 60%. Whether further analysis will show there is a benefit in one particular subgroup, such as those with early stage disease, has yet to be shown. Cyclosporin is associated with a significant incidence of side effects, notably renal impairment (9%) and hypertension (11%). Whether this, together with the cost will preclude its greater use in primary biliary cirrhosis remains to be ascertained. Overall, specific treatment for primary biliary

Neuberger, Lombard, Galbraith

S76

cirrhosis remains far from satisfactory. The number of different agents tried reflects the growing desperation to find a cure for this

10-8

enigmatic disease.

0-6

Prognostic models The increasing use of liver transplantation and X=8 x=5 x=2 its acceptance as a therapeutic modality for endstage disease has led to new emphasis being n0 placed on prognostic models. Because indications for liver transplantation include an estimated survival in the absence of transplantation of one year or less, the ability to define and -1 0 1 2 3 4 5 6 7 8 9 quantify anticipated survival has become increasingly important. The first attempt at Prognostic index quantification of survival was developed from survival probability for surviving two, 2: Estimated Figure data gathered in the 1985 azathioprine trial."9 19). ref five oryears (from six variables: the The model developed included major component was serum bilirubin but other factors included age, serum albumin, histological presence of cirrhosis and central but they have not been confirmed by other cholestasis and whether the patient was treated centres:43 while there is uniform agreement that with azathioprine. By entering the six variables mitochondrial antibodies recur, the histological into a formula (Table III), a prognostic index was findings have suggested this is due to mild derived and from this the one year survival chronic rejection. This controversy has still yet probability or the 50% survival probability can to be resolved but with the increasing number be estimated (Fig 2). The model was further of patients undergoing transplantation and the validated by comparing prognostic data with increasing survival and long follow up, it is actual data in a separate group of placebo anticipated that in the next few years this quespatients, followed as part of the penicillamine tion will be resolved. study. Since then, several other groups have also developed prognostic models using different criteria2021 and it is of interest that the models Immunological abnormalities in primary correlate well. It is of interest that in almost biliary cirrhosis all models, bilirubin, age, and albumin are major In contrast with autoimmune chronic active hepatitis, there is no clear class I or class II HLA components. It must be remembered that all these models association with primary biliary cirrhosis concentrate only on estimates of prognosis although occasional reports have suggested derived from data at referral or diagnosis. It is increased incidence of DRW3 and DRW8. The inappropriate to use these models in a time Liver Unit study did suggest an increased fredependent analysis. Indeed, preliminary data quency of the class III phenotype C4B2 but not analysed by Christensen suggest that different of other class I or II phenotypes." Because these factors are significant prognostic factors in time phenotypes are all found in approximately half the patients with disease, it is unlikely that any dependent models. The ability to give an objective and quantita- of these phenotypes represent a major genetic tive estimate of survival allows an objective influence. It is, however, of interest that C4B2 method of audit of the results of transplantation. has been associated with other chronic diseases Preliminary analysis35 showed that there was a with a presumed viral aetiology and DR3 with significant improvement in survival compared autoimmune diseases. A genetic association may with anticipated survival in the absence of such procedure. This result has been subsequently confirmed by other centres.3738 Prognostic fac- TABLE IlI Prognosis for survival of primary biliary cirrhosis tors for survival after transplantation are not the 2-52xlog serum bilirubin (Vmol/l) same as those indicating prognosis in the absence + 00069xexp ((age -20)/10) - 0-05xserum albumin (g/l) of transplantation (Table IV).39 + 0

~~ 0-4

0-88 (if cirrhosis present) + 0-52 (if not treated with azathioprine)

Does placebo biliary cirrhosis recur after

(a PI of 6 implies a median prognosis of 10 months and of 5 at 22

were reinforced by a subsequent publication42

six months).

months). transplantation? It has been appreciated for many years that the antimitochondrial antibodies persist after orthotopic liver transplantation.-I It is far more TABLE IV Prognosis for surviving six months after controversial, however, as to whether the disease transplantation for primary biliary cirrhosis recurs in the transplanted liver. An initial report serum bilirubin (mmol/l) from the Liver Unit4' shows that in a small + 1-89xlog 1-38 log serum bilirubin/tLmol/l number of patients followed for up to five years, - 1-14 (if transplanted before 1985) 0-92 (if patient has diuretic responsive ascites) there were clinical, serological and histological features of disease recurrence. These findings (a PI of 4 0 implies an 80% and of 5 a 42% probability of surviving

Primary biliary cirrhosis

S77

also explain the rare occurrence of disease in families. Of the serological abnormalities, the association with raised serum IgM, monomeric IgM, total and secretory IgA, and polyclonal IgG is well recognised. None of these findings is specific to primary biliary cirrhosis. Almost diagnostic of the disease is the presence of the antimitochondrial antibody. The antigen is neither liver nor bile duct specific. There is little evidence it is present on the plasma membrane. Inoculation of antigen in a variety of animals induces anti-E2 but without evidence of liver disease. Other autoantibodies have been described in primary biliary cirrhosis: these include nuclear and nucleolar antibodies and anticentromere antibodies which are strongly associated both in primary biliary cirrhosis and in other conditions with sclerodactyly. Immune complexes are increased and there is hypercatabolism of complement.4"47 Nine different antimitochondrial antibodies have been identified and some of these may be detected in a number of diseases, including drug induced disorders and some infective disorders. Primary biliary cirrhosis is associated especially with M2, M4, M8, and M9; it has been suggested that the different subtypes are associated with varying prognosis. M2 recognises an antigen on the inner mitochondrial membrane, the remainder react with antigens on the outer membrane. There is confusing information whether the antigens are present on the surface membrane of cells.48 Four of the major M2 autoantibodies are the E2 components of three 2-oxo-acid dehydrogenase complex and protein x of the pyruvate dehydrogenase complex, an enzyme system present in mitochondrian but encoded in the nucleus. The enzymes are almost universal and are well conserved. Despite the major advances in defining the antigens of the antimitochondrial antibodies system, their significance remains to be established. The universal distribution of pyruvate dehydrogenase complex in all tissues contrasts with the focal nature of disease and the liver. Many bacterial organisms contain pyruvate dehydrogenase complex, therefore, it is possible that a cross reaction may trigger the immune response. Because of the association with other autoimmune diseases and the immune abnormalities, much work has concentrated on examining immunoregulatory cell function. A number of centres have shown that peripheral lymphocytes have impaired suppressor cell activity as tested by concanavalin A induced suppression of mitogen stimulated immunoglobulin synthesis.49 This suppressor cell activity is affected in vitro by corticosteroids, colchicine, and cyclosporin. Monocyte function is also abnormal; possibly a consequence of circulating factors such as immune complexes. Many of these immunological abnormalities, including the presence of antimitochondrial antibodies, are found in a significant proportion of relatives of patients with primary biliary cirrhosis even though they may have no clinical evidence of disease.50 In contrast with these relatively non-specific peripheral immunologic abnormalities, other

work led by the McFarlanes in the Liver Unit has concentrated on organ specific biliary tract antigens." 52 Three antigens, which are speciesspecific high molecular weight carbohydrate or lipid containing protein components, have been identified in normal human hepatic and gall bladder bile. Of these three antigens: one (antigen 1) has not been localised to a particular structure of the liver and is not specific to patients with primary biliary cirrhosis. This may represent an acute phase protein. The second antigen, the canalicular antigen, appears to be derived from the bile biliary canalicular portion of the hepatocellular membrane. The ductular antigen is derived from the epithelial cell lining the intrahepatic bile ducts. Only the latter antigen is specific to the biliary tree. Most patients with primary biliary cirrhosis show cellular immune responses to these antigens and it is of interest that the immune complexes in serum from patients with the disease contain one or more of these antigens. Whether these antigens and their responses in patients with primary biliary cirrhosis represent a primary or secondary immunological event remains far from clear. It is unfortunate that the technical problems in isolating these antigens has precluded further investigations into what remains a novel approach to investigate the disease. Recent studies have suggested that lymphocytes from patients with primary biliary cirrhosis, injected with immune deficient mice can induce a primary biliary cirrhosis-like lesion in the liver.53 If repeated, these results confirm the importance of the search for primary biliary cirrhosis specific antigens on middle sized bile ducts. The future The past 25 years has seen major changes in our understanding of primary biliary cirrhosis. Despite increasing laboratory work revealing yet more abnormalities especially of the immunological system, understanding of the pathogenesis of the disease is still limited. One of the major strengths of the Liver Unit has been its ability to mount and coordinate major intemational studies which have given clear cut answers to the efficacy of different drugs. As yet, no drug has been identified which has a major effect on the progression of the disease. Use of data derived from these trials has allowed generation of prognostic models to optimise the timing of orthotopic liver transplantation. Continuing collaborative studies must be the way forward to confirm or refute therapeutic efficacy of drugs. Primary biliary cirrhosis remains the most common single indication for liver transplantation in Europe and it is to be hoped that therapeutic regimes can be designed on the basis of the understanding derived from work during the past quarter of century to make primary biliary cirrhosis a disease no longer requiring such radical therapy. Figures 1 and 2 are reproduced with kind permission of Gastro-

enterology.

1 Christensen E, Crowe J, Doniach D, et al. Clinical pattern and course of diaease in primary biliary cirrhosis. Gastroenterology 1980; 78: 236-47.

S78

Neuberger, Lombard, Galbraith 2 Crowe J, Christensen E, Doniach D, Popper H, Tygstrup N, Williams R. Presentation of PBC. AmJ Gastroenterol 1985; 80: 466-8. 3 Lucey MR, Neuberger J, Williams R. Primary biliary cirrhosis in men. Gut 1986; 27: 1373-6. 4 Jones EA, Bergasa NV. The pruritus of cholestasis: from bile acids to opiate agonists. Hepatology 1990; 11: 884-7. 5 James OFW, Macklin AF, Watson AJ. Primary biliary cirrhosis - a revised clinical spectrum. Lancet 1981; i: 1278-81. 6 Mitchison HC, Lucey M, Kelly P, Neuberger J, Williams R, James OFW. Symptom development and prognosis in primary biliary cirrhosis. Gastroenterology 1990; 99: 778-84. 7 Stellon AJ, Davies A, Compston J, Williams R. Osteoporosis in chronic cholestatic liver disease. Q J Med 1985; 223: 783-90. 8 Stellon AJ, Webb A, Compston J, Williams R. Low bone turnover state in primary biliary cirrhosis. Hepatology 1987; 7: 137-42. 9 Crowe JP, Christensen E, Butler J, et al. Primary biliary cirrhosis: the prevalence of hypothyroidism and its relationship to thyroid auto-antibodies and Sicca syndrome. Gastroenterology 1980; 79: 1437-41. 10 Murray-Lyon IM, Thompson RPH, Williams R. Scleroderma and primary biliary cirrhosis. Br Med J 1970; ii: 258-9. 11 Clark AK, Galbraith RM, Hamilton EDB, Williams R. Rheumatic disorders in primary biliary cirrhosis. Ann Rheum Dis 1978; 37: 42-7. 12 Crowe JP, Malloy MG, Wells I, et al. Increased C1Q binding and arthritis in primary biliary cirrhosis. Gut 1980; 21: 418-22. 13 Stellon AJ, Williams R. Increased incidence of menstrual abnormalities and hysterectomy proceeding primary biliary cirrhosis. BMJ 1986; 293: 297. 14 Melia WM, Johnson PJ, Neuberger J, Zaman S, Portmann B, Williams R. Hepatocellular carcinoma in primary biliary cirrhosis: detection by alphafetoprotein estimation. Gastroenterology 1984; 87: 660-3. 15 Portmann B, Popper H, Neuberger J, Williams R. Sequential and diagnostic features in primary biliary cirrhosis based on serial histological study in 209 patients. Gastroenterology 1985; 88: 1777-90. 16 Mitchison HC, Bassendine MF, Malcolm AJ, et al. A pilot, double blind controlled one year trial of Prednisolone treatment in primary biliary cirrhosis. Hepatology 1989; 10: 420-9. 17 Coombes B. Prednisolone for primary biliary cirrhosis - good news, bad news. Hepatology 1989; 10: 511-3. 18 Crowe J, Christensen E, Smith M, et al. Azathioprine in primary biliary cirrhosis. A preliminary report of an international trial. Gastroenterology 1980; 78: 1-7. 19 Christensen E, Neuberger JM, Crowe J, Doniach D, Popper H, Williams R. Beneficial effect of Azathioprine and prediction of prognosis in primary biliary cirrhosis. Gastroenterology 1985; 89: 1084-91. 20 Neuberger J, Christensen E, Portmann B, et al. Double-blind controlled trial of d-penicillamine in patients with primary biliary cirrhosis. Gut 1985; 26: 114-9. 21 Roll J, Boyer JL, Berry D, Klatskin G. The prognostic importance of clinical and histological features in asymptomatic and symptomatic primary biliary cirrhosis. N Englj Med 1983; 308: 1-7. 22 Dickson ER, Grambsch PM, Fleming TR, Fisher W, Longworthy N. Prognosis in primary biliary cirrhosis. Hepatology 1989; 10: 1-7. 23 James OFW. D-penicillamine for primary biliary cirrhosis. Gut 1985; 26: 109-13. 24 Bodenheimer HC, Schnaffner F, Sternleib I, et al. A prospective clinical trial of d-penicillamine in the treatment of primary biliary cirrhosis. Hepatology 1985; 5: 1139-94. 25 Kaplan MM, Alling DW, Wolfe HJ, et al. A prospective trial of colchicine for primary biliary cirrhosis. N Engl J Med 1989; 315: 1448-59. 26 Warnes TW, Smith AM, Lee F. A controlled trial ofcolchicine for primary biliary cirrhosis. J Hepatol 1987; 5: 1-7. 27 Bodenheimer H, Schnaffner F, Pezzulo J. Evaluation of colchicine therapy in primary biliary cirrhosis. Gastroenterology 1988; 95: 24-129. 28 James OFW. Ursodeoxycholic acid treatment for chronic liver disease. JHepatol 1990; ii: 5-8.

29 Poupon P, Poupon RE, Calmus Y, et al. Is ursodeoxycholic acid an effective treatment for primary biliary cirrhosis? Lancet 1987; i: 834-6. 30 Hadzyannis Sj, Hadzyannis EJ, Makris A. A randomized controlled trial of ursodeoxycholic acid (UDCA) in primary biliary cirrhosis [Abstract]. Hepatology 1989; 10: 580. 31 Routhier G, Epstein 0, Janossy G, et al. Effects of cyclosporin A on suppressor and induced T lymphocytes in PBC. Lancet 1980; ii: 1223-6. 32 Minuk G, Bohme C, Burgess E, et al. Pilot study of cyclosporin A in patients with symptomatic primary biliary cirrhosis. Gastroenterology 1988; 95: 1356-63. 33 Weisner RH, Ludwig J, Klindon KD, et al. A controlled clinical trial of cyclosporin in the treatment of primary biliary cirrhosis. N EnglJ Med 1990; 322: 1419-29. 34 Hoofnagle J, David GL, Schnafer D, et al. Randomized trial of chlorambucil for primary biliary cirrhosis. Gastroenterology 1986; 91: 1327-4. 35 Kaplan M. Methotrexate treatment of chronic cholestatic disease. Q JMed 1989; 268: 757-61. 36 Lombard M, Portmann R, Neuberger J, et al. Cyclosporin A treatment in primary biliary cirrhosis. [Abstract]. Hepatology 1990; 12: 872. 37 Neuberger J, Altman DG, Christensen E, Tygstrup N, Williams R. Use of a prognostic index in liver transplantation for PBC. Transplantation 1986; 41: 713-6. 38 Markus B, Dickson ER, Grambsch P, et al. Efficacy of liver transplantation in patients with primary biliary cirrhosis. NEnglJ7Med 1989; 320: 1709-13. 39 Neuberger J, Altman DG, Polson R, et al. Prognosis after liver transplantation for primary biliary cirrhosis. Transplantation 1989; 48: 444-7. 40 Smith MGM, Williams R, Doniach D, Caine RY. Effect of orthotopic liver transplantation on serum-auto-antibodies. Lancet 1971; ii: 1006. 41 Neuberger JM, Portmann B, Macdougall BRD, et al. Recurrence of primary biliary cirrhosis after liver transplantation. N EnglJ Med 1982; 306: 1-4. 42 Polson RJ, Portmann B, Neuberger JM, Calne RY, Williams R. Evidence for disease recurrence after liver transplantation for PBC. Gastroenterology 1989; 97: 715-25. 43 Esquivel CU, Van Thiel DH, Bernados A, et al. Liver transplantation for primary biliary cirrhosis. Gastroenterology 1982; 94: 1207-16. 44 Briggs D, Donaldson P, Hayes P, et al. A major histocompatability of complex class III allotype (CB2) associated with PBC. Tissue Antigens 1987; 29: 141-5. 45 James SP, Elson CO, Jones EA, Strober W. Abnormal regulation of immunoglobulin synthesis in primary biliary cirrhosis. Gastroenterology 1980; 79: 242-54. 46 Bernstein R, Neuberger J, Bunn C, Callensen M, Hughes G, Williams R. Diversity of auto-antibodies in primary biliary cirrhosis and chronic active hepatitis. Clin Exp Immunol 1984; 55: 553-60. 47 Wands J, Dienstag J, Bhan AK, Feller ER, Isselbacher KS. Circulating immune complexes and complement activation in primary biliary cirrhosis. N Engl J Med 1978; 298: 223-37. 48 Al-Aghbar MNA, Alexander G, Nouri-Aria K, Neuberger J, Eddleston A, Williams R. In vitro effect of cyclosporin A on immunoglobulin products and concanvalin induced suppression in primary biliary cirrhosis. Gut 1986; 27: 317-23. 49 Van De Water J, Surch C, Leung PSC, et al. Molecular definitions, auto epitopes and enzymatic activities of the mitochondrial auto-antigens of primary biliary cirrhosis. SeminLiverDis 1989; 9:132-7. 50 Galbraith RA, Smith M, Mackenzie RM, et al. High prevalence of sero-immunologic abnormalities in relatives of patients with chronic active hepatitis or primary biliary cirrhosis. N EnglJ3Med 1974; 290: 63-9. 51 McFarlane G, Wojcicka BM, Tsantoulas DL, et al Leucocyte migration inhibition in response to biliary antigens in primary biliary cirrhosis, chronic active hepatitis and other liver diseases. Gastroenterology 1979; 76: 1333-40. 52 Amoroso P, Vergani D, Wojcicka B, Eddleston A, Tee D, Williams R. Identification of biliary antigens in circulating complexes in PBC. Clin Exp Immunol 1980; 42: 95-101. 53 Krams SM, Dorschkino K, Gershwin M. Generation of biliary lesions after transfer of human lymphocytes to severe combined immunodeficient (SCID) mice. J Exp Med 1989; 170:1919-30.

Primary biliary cirrhosis.

Gut Supplement, 1991, S73-S78 S73 Primary biliary cirrhosis J Neuberger, M Lombard, R Galbraith During the last 25 years there have been considera...
1MB Sizes 0 Downloads 0 Views